Background:
Cancer accounts for 22 % of all deaths among children in Europe and is thus the leading
non-traumatic pediatric cause of death. Central nervous system (CNS) tumours constitute 25%
of all childhood cancers, and the majority of these are located in the posterior fossa. One
of the most troublesome late effects after neurosurgery for such a tumour is the cerebellar
mutism syndrome which is seen in up to 25% of the patients. It is characterized by mutism,
hypotonia, ataxia and irritability. The exact aetiology, risk factors, clinical course and
treatment have yet to be identified. The aim of this study is to accomplish that.
Method:
This is an observational prospective multicentre study that will include a minimum of 500
paediatric patients with posterior fossa CNS tumours from the Nordic countries. Additional
countries might be added later once the study is running. The study has started in fall 2014
in 20 centres from 5 Nordic countries. Prior to this a pilot study was performed on 43 Danish
adult patients to validate and fine-tune registration procedure. All patients will be treated
according to local standards, but clinical data will be collected and imaging will be
reviewed centrally.
To calculate the participation rate the annual number of included patients from each country
will be compared to the number of registered patients in the cancer registers of the
respective country and year.
A blood sample for genetic analysis will be collected from all patients. The patients'
neurology and speech functions will be examined both pre-operatively and repeatedly
post-operatively, including recording of standardised speech samples. All data will be
collected trough a, for the purpose developed, online database.
Registration of data. The following data will be registered at the following 5 time points:
1. Preoperatively Hospital and country, and patient related variables (date of birth,
handedness, bilingualism, sex and date of diagnosis). Medical history (Previous
neurological/neuropsychological/ psychiatric problems, comorbidities, previous
operations or other treatment for the tumour, previous regular use of any kind of
medication). Preoperative neurological status will be examined and a language and speech
test will be performed and recorded. If the patient is younger than 2 years a bedside
assessment of the speech will be performed instead of a test. A blood sample for genetic
analysis will be drawn together with the standard blood samples. Alternatively this can
be done at any time during follow-up.
2. Postoperatively within 72 hours of surgery Operation related variables (date, duration
and course of operation, surgical position, surgical approach, and tumour removal
method), complications, technology employed, preoperative hydrocephalus and estimated
completeness of tumour resection.
3. Postoperatively within 1-4 weeks from surgery Approximately 1-2 weeks post-op:
Postoperative language and speech status and for those older than 2 years, a recording
of a speech sample. Neurological examination. Glucocorticoid administration pre-, intra-
or postoperatively plus other medications used to treat the CMS postoperatively and
their effects. Kind(s) of imaging performed on the tumour pre- and postoperatively.
Approximately 4 weeks post-op: Development and treatment of postoperative intracranial
haematoma and hydrocephalus, leakage of cerebrospinal fluid and need for ventilator.
4. Postoperatively at about 2 months from surgery Postoperative development of CMS,
detailed survey of the status of CMS in those affected including recording of speech
sample and neurological examination. Medications used to treat the CMS since last
registration and their effects.
5. Postoperatively at about 12 months from surgery Language and neurological status
including a speech sample for those older than 2 years. Medications used to treat the
CMS since last registration and their effects. Other anti-cancer treatment given
(chemotherapy and/or radiotherapy). Results from the pathology department regarding the
kind of tumour histology and genetics. Registration of whether neuropsychological
assessment(s) have been performed. Kind(s) of imaging performed on the tumour since 1st
follow-up. Copies of the MRIs and descriptions performed pre-op, right after the
operation and approximately. 12 months post-op are obtained.
All registered data will be examined by a third party who will check for missing data or
misentries in order to ensure a high quality of the data. In case of missing data or
misentries the third party will contact the person who made the registration.
The database is administered by the children's cancer epidemiology group (CCEG) at Karolinska
Hospital in Stockholm, which is also responsible for the Swedish children's brain tumour
registry and the leukaemia database for the Nordic Society of Pediatric Hematology and
Oncology (NOPHO)
Other courses:
In case of acute surgery, coma etc. information about the study and the offer to participate
can be given within 7 days from the operation. In these cases preoperative data about the
patient can be obtained from the patient's medical record and/or from the parents, but a
preoperative speech sample cannot be performed. These patients will not be included in the
analysis of whether and how preoperative speech and language status affects the risk of
developing CMS, but will be included in all the other study analyses. Speech samples will be
performed postoperatively in exactly the same manner as in patients that were included before
operation to able to monitor the patient's speech postoperatively and register signs of the
CMS.
Should the patient have posterior fossa tumour surgery performed again during the 12 months
follow-up period, the patient will re-start the follow-up programme from that date. A
separate pre-op and all the post-op registrations will be performed again, and used in the
analysis of risk of first versus second or further surgery. New genetic blood samples will
not be necessary in these cases. Should the patient have posterior fossa tumour surgery
performed after the last 12 months follow-up, the patient will be offered to participate in
the study again and a new consent will have to be obtained. New genetic blood samples will
not be necessary in these cases.
If the patient leaves the study for any reason before the follow-up 12 months post-op, a
separate form will be filled out explaining why the patient left.
If the patient turns 18 while included in the study, a new consent to participate will have
to be given by the patient him- /herself.
Blood samples and analyses:
As soon as a patient has been registered, the study centre will request a 2 ml anticoagulated
blood sample for genetic profiling (Single-Nucleotide Polymorphism (SNP) analysis).
The investigators will use a newly developed single nucleotide polymorphism (SNP) sequencing
strategy that allows cost-effective mapping of 25-30.000 genetic polymorphisms within
biological domains that could potentially be linked to the development of CMS (e.g.
inflammation, vascularization, blood-brain-barrier markers, and apolipoprotein E and other
lipoprotein pathway genes). The genetic data will be linked to the clinical data to identify
genetic variants associated with the risk of CMS or the course of CMS. Specifically the
investigators will map all SNPs in all genes that are known or are likely to be linked to
these pathways, including mRNA binding sites and first order protein-protein interactions.
Rather than expecting large effects of single SNPs, the strategy of this approach is to use
front-line bioinformatics and pathway analyses to explore the additive effect of numerous
SNPs involved in the same biological pathway. This will identify high-impact pathways,
although with individual low-impact SNPs. The results obtained could guide future therapeutic
approaches to CMS.
If the custodial parents do not consent to their child contributing a blood sample to the
study, the child may still participate in the study, albeit not in the part involving genetic
analysis. The samples as well as the rest of the study data will be protected under the Act
on Processing of Personal Data and the Act on the Health Act.
All MRIs are analyzed by neuroradiologists with respect to tumour resection and
neuroradiographic signs associated with CMS.
All speech recording are analyzed by speech therapists with respect to signs associated with
CMS. The results of neuropsychological tests that may have been performed routinely will be
separately obtained and analyzed.
Power calculations:
For the surgical hypothesis, assuming that 35% of patients are operated with an approach that
has a lower risk of CMS (assumed to be 10%) and the remaining 65% of patients are operated
using other approaches that carries a 20% risk of CMS, the investigators will with 80% power
be able to identify a difference at a 5% significance level if the investigators include a
total of 450 patients.
For the genetic analysis, several pathways and SNP-profiles will be explored with appropriate
adjustments for multiple comparisons. Multiple SNPs will due to randomness be found to be
related with the risk of CMS. Their true biological significance will subsequently be
validated through internal validation, as the investigators will explore if other SNPs in the
same biological pathway, e.g. SNPs in the same genes but not the same haplotype or SNPs that
affect coding or regulatory regions in the identified risk-related genes are more
significantly associated with risk of CMS than randomly selected SNPs. Furthermore, the
genes/SNPs will be explored by bioinformatic predictions of the impact of the SNPs on
protein-folding, binding affinity etc. Once such high-risk SNPs/genes/pathways have been
identified and published, the investigators will attempt to have them confirmed in
independent patient cohorts from Europe or the US. Based on a projected overall risk of CMS
of 20%, a frequency of a specific SNP (or SNP-profile) of 30%, and a projected doubled risk
of CMS in the group that harbour the SNP (or SNP-profile), the investigators will with 90%
power be able to identify such a genetic predisposition at a 5% significance level, if a
total of 343 patients are included in the study. Thus, the study has sufficient power.
To analyze the effect of the above mentioned variables (surgical method, administration of
corticosteroid, handedness etc.) on the risk of developing CMS the investigators will perform
univariate and multivariate regression analyses as well as standard descriptive analyses.
These analyses will be performed using R.
Discussion:
The study will be the largest prospective multicenter study on cerebellar mutism syndrome to
date, and the first one of its kind to systematically gather detailed information about 1)
the surgical approaches least likely to cause the syndrome, 2) how the syndrome is best
treated, 3) the role of genetics and 4) differences in incidence and clinical course of the
syndrome for different patients.
The ultimate aim of the study is to reduce the incidence and improve the treatment of
cerebellar mutism syndrome and lead to harmonization of the treatment of CNS tumour patients
across the Nordic countries.
